All this is basically troubleshooting for a network. The best-case reliable, scalable, usable way to do networking/sharing on such a hodgepodge of user groups/computers/etc. is setting up a server/domain/network shares/connect users. I know people don't always want to do this because of server/expense/blah but it is just too much of a pain, and it is too unreliable to setup local workgroups. If you've got a few people.. fine.. if not, talk to Brenda Knavel 6310 and it is something like $130 per workstation per year to have file/printer shares, backup, security and all the other good stuff. Or if that doesn't work for you just buy a server and set it all up. Human Resource, Provost, and the other departments have no problem sharing!
The only reason I'm saying that is all this information below is just troubleshooting because we don't have a nice clean network. The only reason you have to even worry about this stuff is because network neighborhood is meant for small networks.
That is only a suggestion for troubleshooting something better, BUT, I'm now going to hopefully give you some info you can use.
#1 - You don't say you're using network neighborhood so I'll troubleshoot trying start/run \\machinename
If you're static and have a wins you're at risk that the number is old, so just pop it over the DHCP and try that. The current WINS should be 137.229.18.32
Hybrid node searches in this order:
1 Name Cache (any names that have been resolved already and put in a 'cache' for 10 minutes) 2 WINS (3 attempts) 3 Broadcast 4 Lmhosts file 5 Hosts file 6 Dns server - request sent at 5, 10, 20, 40, 5, 10, and 20 seconds
So make sure node type and then try to ping to test.
Try nbtstat -c to see what is in the cache. Try and ping (computer) or \\(computer) and see what has been resolved.
Try nbtstat -R to reload the cache.
#2 - If you are using network neighborhood (hopefully not, it's useless and unreliable!) I'm assuming you are at least 2k if not xp... 9x/ME have their own problems connecting.
If you look in your start/run eventvwr look for MRxSmb/browser in source with an event ID of something like 8003/8015. What this is going to tell you is who the master browser is. Most likely the way you explain the situation your problem is with this computer! If you look through the logs, are their a lot of elections for masterbrowser? Does it get changed a lot?
When you go into network neighborhood you are asking the master browser to display the network information. In fact, I believe that when this message comes up, it starts to try to elect a new master browser within the workgroup.
The master browser is elected by windows version 3.11 95 NT 2000 XP NT server 2000 Server Linux (if configured badly/correctly)
Linux fits in there somewhere and actually one of the 'evils (or goods)' of linux is that it can force an election and take control as master browser.
So if you have a mix of machines with the newest being 2000 boxes than only those 2000 boxes can be master browser. Immediately if an XP or higher machine comes on the workgroup it will try to re-elect as master browser (and win!)
Whenever a new browser wins it will actually stop the network neighborhood (most likely) from working while it captures all the machines. Machines can take upward of 15 minutes to capture and 51 minutes to delete from the master browser list.
If you have everybody do an nbtstat -a (computername) then you can find out who is the masterbrowser
Name Type Status --------------------------------------------- FXJCS <20> UNIQUE Registered FXJCS <00> UNIQUE Registered UAF <00> GROUP Registered UAF <1e> GROUP Registered UAF <1d> UNIQUE Registered ..__MSBROWSE__.<01> GROUP Registered
MAC Address = 00-50-56-C0-00-01
If you have the ..__MSBROWSE__.<01> then you are the masterbrowser. You can also just type in name of any machine. So if you want to play around you can type in other names like nbtstat -a millennium.
Since NetBEUI is non-routable through subnets, you are contesting with everyone that is in your workgroup in the subnet. You might be looking around for other people that could be causing this problem! For example, if there is a Linux box with Samba on it that is taking over the workgroup network, you should ask them nicely to stop.
You can also just go to start/search or find computers and that will bypass the master browser.
#3 - Try enable Netbios over TCP/IP, maybe being on DHCP disabled or enabled. If you don't have anything less than 2k you could try disabled which would use SMB traffic, but older OS versions need NBT traffic so you'll have to enable it.
#4 - If you've got a firewall, Netbios over tcp/ip uses
Nbname 137/udp Nbname 137/tcp Nbdatagram 138/udp Nbsession 139/tcp
Netbios-less SMB traffic uses port 445 tcp/udp
#5 - Workaround - Just setup one machine with all the shares and connect to that so you don't have to worry about any other computers through the network neighborhood.
#6 - Another workaround - Get shortcuts for all the shares that you have on the network and e-mail that to your users so they just have a makeshift 'network neighborhood' that is totally reliable. Basically you're doing all the master browser manually, but once you set it up it will be much more reliable than what you've currently got.
Good luck.